Pricing

Access to Lite versions of Chat, Code, and Vision models.

100,000 tokens per month.

Basic API access.

Email-based technical support.

Access to Lite models up to 7B parameters.

500,000 tokens per month.

Enhanced API speeds.

Standard support.

Access to Turbo-mode models up to 7B.

1 million tokens per month.

Basic inference configuration.

Turbo access to models up to 13B parameters.

1.5 million tokens per month.

Advanced API request settings.

Lite access to models up to 13B parameters.

3 million tokens per month.

Premium SLA support (response within 24 hours).

Turbo access to models up to 13B parameters.

5 million tokens per month.

Priority support.

5 individual API keys.

10 million tokens per month.

User management features.

10 individual API keys.

20 million tokens per month.

Full access to all models up to 13B in Turbo mode.

Dedicated server powered by NVIDIA A100/L4.

Models up to 13B parameters.

Unlimited local queries.

Dedicated server.

Full model customization access.

Support for custom API integrations.

Fine-tuning for models up to 7B parameters.

Hyperparameter configuration.

Dedicated support specialist.

Fine-tuning for models up to 13B parameters.

Support for custom datasets.

Guided setup and tuning process.

Dedicated inference cluster.

Private API keys.

Turbo model access.

Private server hosted in an isolated data center.

Full infrastructure control.

High-availability SLA.

Custom-built models.

Flexible licensing.

Multi-instance support.

Fine-tuning for any model up to 70B parameters.

Custom dataset integration.

Premium technical support.

Fully isolated cloud environment for models and data.

Complete operational autonomy.

Enterprise-grade API access.

Clusters powered by NVIDIA A100/H100 GPUs.

Unlimited model training.

Access to experimental features.

Full-stack training, inference, and private APIs.

24/7 personal account manager.

Priority access to all upcoming innovations.